Pulsed thermal annealing enables switching of chiral antiferromagnetic order with a sub-millitesla field in Mn$_3$Sn

This study demonstrates that combining pulsed thermal annealing above the Néel temperature with a sub-millitesla magnetic field enables reliable and reproducible switching of the chiral antiferromagnetic order in Mn$_3$Sn by thermally softening magnetic anisotropy to near-zero energy barriers.

The Gist

Imagine you have a very stubborn, tiny compass inside a piece of metal. This isn't a normal compass; it's made of atoms that are arranged in a special, twisted pattern (called "chiral antiferromagnetic order"). In a normal compass, the needle points North because of a strong magnetic pull. But in this special metal, the "needle" is locked in place by a heavy, invisible door that won't let it move unless you push with a lot of force.

For a long time, scientists tried to flip this needle using only magnetic pushes. They found they needed a surprisingly strong push to get the needle to turn around.

The Big Discovery: The "Heat Key"
This paper, by researchers at Huazhong University of Science and Technology, reveals a much easier way to flip that needle. They discovered that if you briefly warm up the metal, the heavy door disappears.

Think of it like this:

• The Metal: A block of Mn3Sn (a special type of metal).
• The Needle: The internal magnetic direction of the atoms.
• The Door: The "magnetic anisotropy," which is the energy holding the needle in place.
• The Heat: A quick pulse of warmth.

When the researchers heated the metal just a little bit past a specific "tipping point" temperature (called the Néel temperature, around 425 Kelvin or 152°C), the door to the needle's movement swung wide open. The atoms became "soft" and loose, no longer caring which way they pointed.

The Tiny Nudge
Here is the magic part: Once the door is open because of the heat, you don't need a strong push to move the needle. You only need a tiny, almost invisible nudge.

The researchers showed that after heating the metal, they could flip the needle's direction using a magnetic field so weak it was barely there—about 0.1 millitesla. To put that in perspective, that is roughly 1/100th the strength of a standard fridge magnet. Without the heat, that tiny nudge would have done absolutely nothing. But with the heat, the needle instantly snapped into the new direction as the metal cooled back down.

The Experiment
The team built a special setup where they could heat the metal with a separate heater (so the electricity used for heating didn't interfere with the measurements) and then let it cool down while applying that tiny magnetic nudge.

• Heating: They warmed the metal to 438°C (above the tipping point) for 10 seconds.
• Cooling: As it cooled, they applied a tiny magnetic field pointing either left or right.
• Result: The metal's internal magnetism flipped completely to match that tiny field. If they pointed the field left, the needle went left. If right, it went right.

They also proved that if they heated the metal but didn't cross that tipping point (staying just below it), the door didn't open fully, and the tiny nudge couldn't flip the needle. The heat had to be strong enough to erase the old direction completely.

Why This Matters for Future Gadgets
The paper suggests that in future computer memory chips (which use these materials to store data), we shouldn't fight against the heat. Usually, engineers try to stop devices from getting hot because heat is seen as a problem.

This paper argues that heat is actually a helper.

• Old Way: Try to push the needle with a huge force (high energy) to break the door.
• New Way: Use a burst of heat to unlock the door, then use a tiny, low-energy push to set the direction.

The researchers even provided a simple math recipe to help engineers figure out how much a tiny chip will heat up when a pulse of electricity runs through it. This helps them design devices that intentionally use this "heat-assisted" method to save energy and switch data faster.

In Summary
The paper shows that to flip the magnetic switch in Mn3Sn, you don't need a sledgehammer. You just need to briefly warm it up to unlock it, and then a whisper of a magnetic field is enough to set it in the new direction. It turns out that heat and magnetic forces are partners, not enemies, in building the next generation of fast, efficient memory.

Technical Summary

Technical Summary: Pulsed Thermal Annealing Enables Switching of Chiral Antiferromagnetic Order in Mn3Sn

Problem Statement
The manipulation of antiferromagnetic (AFM) order is a central objective in modern spintronics, offering advantages such as ultrafast dynamics, absence of stray fields, and robustness against external perturbations. Among AFM candidates, the chiral AFM Mn3Sn is particularly promising due to its non-collinear 120° spin structure on a kagome lattice, which generates a cluster magnetic octupole order. This order breaks time-reversal symmetry, producing a large anomalous Hall effect (AHE) at room temperature without net magnetization. While the octupole vector can be electrically switched, a critical aspect of this process—thermal softening—has often been overlooked. In many spin-torque-driven switching experiments, the high current densities required ($10^7$ A/cm$^2$) inevitably generate significant Joule heating. As the temperature approaches the Néel temperature ($T_N \approx 425$ K), the magnetic anisotropy that pins the octupole order weakens. Above $T_N$, the magnetic order is erased, and the reorientation barrier vanishes. Despite the physical clarity of this mechanism, its specific role in enabling switching by weak fields has not been fully isolated or discussed in recent literature, particularly in studies focusing on spin–orbit torque (SOT).

Methodology
To isolate the role of thermal softening from spin-torque effects, the authors conducted controlled experiments on high-quality bulk single crystals of Mn3Sn.

• Sample Preparation: Mn3Sn single crystals were grown using the Bridgman–Stockbarger method and oriented via Laue diffraction.
• Experimental Setup: The sample was mounted on a resistive heater chip using insulating glue, ensuring that the heating current did not pass through the sample itself. An E-type thermocouple attached directly to the sample monitored the temperature, while a copper cylinder acted as a heat sink for rapid cooling.
• Measurement Protocol: Hall resistivity ($\rho_{zy}$) was measured using a standard four-probe configuration with in-plane magnetic fields and out-of-plane currents. The experiments were performed in a Physical Property Measurement System (PPMS) at a base temperature of 250 K.
• Pulsed Thermal Annealing: The core experimental procedure involved heating the sample to 438 K (above $T_N$) for 10 seconds to ensure thermal equilibrium and erase the magnetic order. The sample was then cooled in the presence of a small, static external magnetic field (ranging from 0.1 mT to 0.6 mT). This approach decoupled thermal effects from spin-torque effects, allowing the authors to determine the minimum field required to set the final magnetic state solely through thermal softening.

Key Results

1. Threshold Field Dependence: Systematic measurements of the AHE hysteresis loops at various temperatures revealed that the threshold field ($B_0$) required to switch the octupole order decreases monotonically as temperature increases. $B_0$ extrapolates to zero at $T_N \approx 425$ K, confirming that thermal energy progressively reduces the switching energy barrier until it vanishes.
2. Switching with Sub-millitesla Fields: Pulsed thermal annealing experiments demonstrated that heating the sample above $T_N$ followed by cooling in a tiny magnetic field enables reliable switching of the magnetic octupole order.
   • A cooling field of $\pm 0.4$ mT resulted in complete suppression of the order during heating and full recovery with the sign matching the cooling field.
   • The switching ratio curve is extremely sharp: a field of $\pm 0.3$ mT yields nearly full saturation (>90% of maximum AHE), and even $\pm 0.1$ mT achieves a switching ratio of $\sim 70\%$.
   • Crucially, the switching process showed no hysteresis regarding the field cycling history; the final state depended solely on the sign of the field applied during cooling.
3. Necessity of Crossing $T_N$: Comparative experiments showed that annealing at 409 K (below $T_N$) with a 0.2 mT cooling field failed to switch the order. Instead, the original orientation was only partially attenuated. This confirms that crossing $T_N$ is essential to completely erase the magnetic order and allow a weak field to dictate the new orientation.
4. Thermal Modeling: The authors provided a simple analytical model to estimate transient temperature rises ($\Delta T$) in nanoscale devices under current pulses.
   • For continuous thin films, $\Delta T$ is limited by bulk resistivity, requiring very high current densities ($J \gtrsim 10^8$ A/cm$^2$) or long pulses to reach $T_N$.
   • For nanoscale devices (e.g., nanopillars), where contact/tunnel resistance dominates, the model predicts that moderate current densities ($J \sim 10^7$ A/cm$^2$) can generate $\Delta T \approx 1000$ K, easily reaching $T_N$.

Significance and Claims
The paper argues that thermal softening is not an undesirable side effect of high-current operation but a crucial, enabling mechanism for switching chiral antiferromagnets.

• Mechanism Clarification: The work establishes that heating above $T_N$ removes the magnetic anisotropy barrier, lowering the energy required for reorientation to nearly zero. This allows extremely weak directional fields (such as the effective field from spin–orbit torque) to determine the final magnetic state during cooling.
• Reinterpretation of SOT: The authors suggest that in thin-film SOT devices, the spin current should be viewed not as the sole driver fighting a full anisotropy barrier, but as the directional bias in a heat-assisted process. This perspective helps reconcile why some SOT experiments required external fields (insufficient heating) while others succeeded with longer pulses or higher currents (effective thermal softening).
• Design Guidance: By providing a thermal model, the paper offers practical guidelines for device design. It suggests that future devices should intentionally engineer thermal profiles to ensure the device core reaches $T_N$ during write pulses. This approach could enable low-power, fast, and reliable all-electrical switching, as the SOT effective field only needs to provide a modest bias rather than overcome the full anisotropy barrier.
• Conclusion: The authors conclude that ignoring thermal effects risks misunderstanding the switching mechanism. In chiral antiferromagnets like Mn3Sn, heat and spin act as partners rather than adversaries, and future work must account for both to achieve optimal device performance.